Environmental and Efficiency Analysis of Simulated Application of The

energies

Article Environmental and Eﬃciency Analysis of Simulated Application of the Solid Oxide Fuel Cell Co-Generation System in a Dormitory Building

Han Chang * and In-Hee Lee *

Department of Architecture, Pusan National University, Busan 46241, Korea * Correspondence: [email protected] (H.C.); [email protected] (I.-H.L.)

Received: 23 August 2019; Accepted: 11 October 2019; Published: 15 October 2019

Abstract: The problem of air pollution in Korea has become progressively more serious in recent years. Since electricity is advertised as clean energy, some newly developed buildings in Korea are using only electricity for all energy needs. In this research, the annual amount of air pollution attributable to energy under the traditional method in a dormitory building, which is supplying both natural gas and electricity to the building, was compared with the annual amount of air pollution attributable to supplying only electricity. The results showed that the building using only electricity emits much more air pollution than the building using electricity and natural gas together. Under the traditional method of energy supply, a residential solid oxide fuel cell cogeneration system (SOFC–CGS) for minimizing environmental pollution of the building was simulated. Furthermore, as a high load factor could lead to high eﬃciency of the SOFC–CGS, sharing of the SOFC–CGS by multi-households could increase its eﬃciency. Finally, the environmental pollution from using one system in one household was compared with that from sharing one system by multi-households. The results showed that the environmental pollution from sharing the system was relatively higher but still similar to that when using one system in one household.

Keywords: environmental impact; device eﬃciency; air pollutant; multi-households; solid oxide fuel cell cogeneration system

1. Introduction

1.1. Research Background Electricity as non-combusted energy is advertised to the public as clean energy. Since the price of natural gas and the price of electricity for residential use are similar [1], in the areas of architecture and construction, some real estate developers have been developing new buildings using only electricity for power. However, as a secondary energy resource, electricity is either clean or not clean depending on the electricity generation procedure. In Korea, 39% of electricity is generated from burning coal and only 4% is generated from renewable resources as seen in Figure1[2].

Energies 2019, 12, 3893; doi:10.3390/en12203893 www.mdpi.com/journal/energies Energies 2019, 12, 3893 2 of 20 Energies 2019, 12, x FOR PEER REVIEW 2 of 20

6% 1% 4% Electricity generation by using nuclear energy 31% Electricity generation by combusting coal 19% Electricity generation by using LNG

Electricity generation by using renewable resources

Electricity generation by other fuel resources

Electricity generation by hydraulic power 39% Figure 1. Electricity generation system in Korea.

The traditional methodmethod ofof supplyingsupplying energy energy to to a a building building in in Korea Korea is mainlyis mainly using using natural natural gas gas for forheating heating and and hot hot water water and and using using electricity electricity for for lighting, lighting, cooling, cooling, and and equipment. equipment. AA recentrecent method of supplying energy to a building is using only electricity for all energy needsneeds inin thethe building.building. This research compared the annual amount of air pollution attributable to energy under the traditional method in a dormitory building, which is supplying both natural gas and electricity to the building and the annual amount of air pollution attributable to supplying only electricity.electricity. A solid oxide fuel cell (SOFC) is a clean energy devicedevice for generatinggenerating electricity by consuming hydrogen or natural gas. By By utilizing utilizing chemical chemical interaction interaction in in the the cell cell stack during the process of electricity generation in SOFC, there is no environm environmentalental pollution. However, there is a large amount of heat dissipation during SOFC operation. In In order order to to improve the efficiency efficiency of SOFC, the SOFC cogeneration system (SOFC–CGS) waswas developeddeveloped toto collectcollect andand useuse the heat loss. The heat heat dissipation dissipation generated from SOFC is collected by residential SOFC–CGS and supplied to a hot water tank for hot water demand of a household. A fuel cell as a clean energy system is widely studied for residential use. In In recent recent years, years, various various researches researches were were done done about about optimizing optimizing the theoperation operation efficiency efficiency of a offuel a cell.fuel cell.Some Some scholars scholars proposed proposed efficiency efficiency optimiza optimizationtion methods methods of ofa afuel fuel cell cell through through improving working principles of electronicelectronic components.components. Wa Wangng et al. proposed proposed a a novel stack and converter interpreted design scheme to improveimprove thethe converterconverter eefficiencyfficiency [[3].3]. A micromicro tri-generation system could increase the system eefficiencyfficiency toto overover 90%90% [[4].4]. An energy management system could also achieve cost cost reduction, reduction, CO CO2 2mitigation,mitigation, and and energy energy consumption consumption [5]. [ 5Some]. Some scholars scholars aim aimto find to finda way a wayto improve to improve the efficiency the efficiency of a fuel of a cell fuel from cell fromanalysis analysis of energy of energy demand, demand, energy energy distribution, distribution, and energyand energy supplication supplication of a offuel a fuelcell. cell.Adam Adam et al. et pr al.esents presents a multi-objective a multi-objective modeling modeling approach approach that allowsthat allows an optimized an optimized design design of micro-combined of micro-combined heat heat and and power power (CHP) (CHP) systems systems considering considering the source, distribution and emission requirements in un unisonison to to achieve achieve more more efficient efficient whole whole systems systems [6]. [6]. Coordination between utilizationutilization factorfactor ofof SOFCSOFC andand featuresfeatures ofof aa DCDC/AC/AC (Direct(Direct current current//AlternatingAlternating currentcurrent)) inverter inverter is is also also an an efficient efficient control control strategy for SOFC operation [[7].7]. Some researchers have combined a photovoltaic, battery, battery, or heat pump wi withth a fuel cell to satisfy the energy needs by a household while reducing cost and CO2 emission [8–10]. [8–10]. From From a a wi widerder energy supply perspective, SOFC was considered toto supplysupply energyenergy forfor bothboth householdshouseholds and and EV EV (Electric (Electric vehicle)vehicle) to to improve the efficiencyefficiency andand energyenergy supply supply capacity capacity of of SOFC. SOFC. Using Using a hydrogena hydrogen supply supply system system for afor residential a residential fuel fuelcell tocell provide to provide energy energy for fuel for cell fuel vehicles cell vehicles is also ais method also a method to solve theto solve problem the ofproblem lacking of hydrogen lacking hydrogenstations for stations fuel cell for vehicles fuel cell [11 ,12vehicles]. Field [11,12]. experimental Field studiesexperimental demonstrated studies thatdemonstrated a residential that fuel a residentialcell could significantlyfuel cell could reduce significantly the CO2 reduceemissions the andCO2 primaryemissions energy and primary consumption energy [ 13consumption]. Accurate [13].mathematical Accurate modelsmathematical were developed models were to analyze developed the performanceto analyze the of SOFC.performance A 1D dynamicof SOFC. modelA 1D wasdynamic built formodel studying was systembuilt for integration studying and system developing integration adequate and energy developing management adequate and controlenergy managementstrategies of SOFCand control systems strategies [14]. Gallo of SOFC et al. developedsystems [14]. a model Gallo ofet aal. non-conventional developed a model SOFC of systema non- conventionalby applying aSOFC lumped system energy by applying balance toa lumped each component. energy balance This to model each comp couldonent. efficiently This model increase could the efficientlyheat exchanges increase inside the theheat system. exchanges This inside model the is verified system. by This experiments model is verified and applicable by experiments to numerous and applicablelayouts [15 to]. numerous According layouts to reference [15]. According [16], a fuel to cell reference is mainly [16], used a fuel in America,cell is mainly Europe, used South in America, Korea Europe, South Korea and Japan worldwide. From environmental aspect, analysis of CO2 reduction by a fuel cell was also shown in recent researches from Europe, Japan and South Korea. Fuel cell for

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 3 of 20

and Japan worldwide. From environmental aspect, analysis of CO2 reduction by a fuel cell was also shown in recent researches from Europe, Japan and South Korea. Fuel cell for EV is mainly applied and investigated in Korea. Environmental impact of CO2 reduction and emission by fuel cell EV is conducted [17,18]. In order to encourage the utilization residential fuel cell in households in Korea, Lim et al. did a study in the view of CO2 emissions reduction by applying residential fuel cell [19]. European researchers did several analyses of applying SOFC in the industrial area to reduce CO2 emission [20,21]. On the other hand, a residential fuel cell is mainly studied in Japan in recent years. The CO2 emission density of a public power plant was used to calculate the amount of CO2 emission and CO2 reduction can be calculated by a simulating application of a fuel cell in a household [22–24]. The same calculation method was used for environmental analysis in this research. From the energy demand oriented research, Ozawa et al. indicates that SOFC–CHP systems can drastically reduce greenhouse gas (GHG) emissions from a particularly small-sized household [24]. Another aspect studied in this research is the efficient operation of SOFC–CGS in small-sized households in a dormitory building by using a exist model of SOFC–CGS from Japan [25–27]. This study analyzed efficiency improvement and environmental impact of operating SOFC–CGS in a dormitory building in Korea by simulating SOFC–CGS operation in multi-households. Under the traditional energy supply method, solid oxide fuel cell cogeneration system (SOFC–CGS) was attempted to simulate applying in the target building of this research. The target building was a dormitory building near a university in Korea. In Korea, the dormitories inside the university campus usually cannot accommodate all of the students registered in the university. Thus, most of the students have to rent a small room for living off campus. Therefore, there are a lot of dormitory buildings near universities in Korea. The dormitory building usually contains several households in each floor and area of one household is around 20 to 30 m2, which can only accommodate one or two students. The target building is near a university campus in Busan. Specifications of the target building can be checked in Table1 in Section 2.1. SOFC–CGS is a clean energy system that does not release any air pollutants when it generates electricity [6,14,24,28]. The device efficiency and environmental impact were analyzed under the conditions of applying one system to one household and applying one system to multi-households to identify the most appropriate way to operate SOFC–CGS in the building. The model of SOFC–CGS is introduced in Section4. Four different types of energy supply were considered in this research. The first type was supplying only electricity to the building, and the second type was supplying natural gas and electricity together. According to the analysis result, the second energy supply type emits a relative smaller amount of air pollutants than the first type and the efficiency of using natural gas directly by combustion is much higher than using natural gas by SOFC for heating. Therefore, the third type was applying one SOFC–CGS to one household under the second energy supply type, and the fourth type was applying one SOFC–CGS system to multi-households under the second energy supply type.

Table 1. Building speciﬁcations.

Parameter Speciﬁc Characteristic Area 172.3 m2 Structure Steel–concrete Envelop Concrete Aboveground Stories 4 Underground Stories 0 Story height 2.6 m Height 10.4 m Width 12.6 m Length 14.5 m Service life 50 years Energies 2019, 12, 3893 4 of 20

1.2. Research Methodology This study mainly used the commercial software Design Builder (Version 4.5, DesignBuilder Software Ltd, London, UK) to predict energy consumption of the building. In the second part of this research, to simulate operating the SOFC–CGS in the building, a calibrated SOFC–CGS mathematical model was programmed using Visual Basic.NET. The research target building was modeled using Design Builder, and the one-year hourly energy consumption of the building was also predicted using this software. Design Builder as a calibrated software is made in England. This software was mainly used for energy analysis of the building. The calculation model of this software is EnergyPlus (Version 8.2.0, Department of Energy, Washington, USA), which is a calculation engine invented in America for analyzing the heating, cooling, lighting, ventilation, etc., of a building. There are several calculation engines for energy consumption prediction from which DOE-2 (Version 2.1E, Department of Energy, Washington, USA), DeST (Version 2.0, Tsinghua University, Beijing, China), and EnergyPlus are mainly used for the analysis of the energy consumption of buildings. Xin et al. compared the simulation performance among these three diﬀerent calculation engines. Compared with EnergyPlus and DeST, DOE-2Energies 2019 is limited, 12, x FOR in PEER the basicREVIEW assumption of the heating, ventilation and air conditioning (HVAC)4 of 20 system calculations [29]. Therefore, the commercial software Design Builder with the calculation engineresearch EnergyPlus target building was chosen were calculated for doing this and research. compared The under air pollutants the conditions of the researchof using targetthe traditional building weremethod calculated of supplying and comparedenergy to the under building the conditions and new method of using of the supplying traditional energy method to the of building supplying by energyreferring to to the the building air pollutant and new emission method densities of supplying of electricity energy and to the natural building gas byin referringKorea. Finally, to the the air pollutantenvironmental emission impact densities and ofefficiency electricity analyses and natural of operating gas in Korea. the SOFC–CGS Finally, the environmentalin the building impact were andconducted. eﬃciency analyses of operating the SOFC–CGS in the building were conducted.

2.2. Energy and Environmental Analysis ofof thethe DormitoryDormitory BuildingBuilding

2.1. Modeling of the Dormitory Building 2.1. Modeling of the Dormitory Building There are a lot of dormitories for accommodating students near the campus of universities in There are a lot of dormitories for accommodating students near the campus of universities in Korea. Each room in the buildings typically can accommodate one or two students. Most of the Korea. Each room in the buildings typically can accommodate one or two students. Most of the dormitory buildings are constructed so that there is a supply of natural gas and electricity to the dormitory buildings are constructed so that there is a supply of natural gas and electricity to the building. However, recently, some newly developed dormitory buildings have been constructed to use building. However, recently, some newly developed dormitory buildings have been constructed to only electricity for all of the energy needs. One of these newly developed buildings was selected for use only electricity for all of the energy needs. One of these newly developed buildings was selected this research. To predict the energy consumption of the building, a model of the building was made for this research. To predict the energy consumption of the building, a model of the building was using Design Builder, and the 3D model and conﬁguration are shown in Figure2. made using Design Builder, and the 3D model and configuration are shown in Figure 2.

(a) (b)

FigureFigure 2.2. 3D modelmodel andand conﬁgurationconfiguration ofof thethe dormitorydormitory building.building. ((aa)) 3D3D model;model; ((bb)) Conﬁguration.Configuration.

AccordingAccording toto thethe energyenergy andand environmentalenvironmental analysisanalysis ofof thethe dormitorydormitory buildingbuilding inin aa previousprevious studystudy [[30],30], toto predict the energy consumptionconsumption of the building precisely, weatherweather datadata of Busan, Korea, werewere acquired,acquired, andand thethe occupancyoccupancy scheduleschedule ofof didifferentﬀerent partitionspartitions ofof thethe buildingbuilding werewere referredreferred fromfrom ASHRAEASHRAE standardstandard 90.2,90.2, 20102010 [[31].31]. Basic weather datadata were listedlisted inin TableTable2 2.. Moreover,Moreover, the the energy energy consumption standard of a building in Korea was referred from the Korean Energy Agency to build the model. According to this standard, heating, cooling, resident density, lighting density, and equipment density were set as in Table 3 [32]. According to Korean Construction Law, the construction thermal specifications of external wall, internal wall, roof, and floor were set, as listed Table 3 [32–34]. Since this building uses only electricity, cooling and heating were both set from electricity. Table 3 briefly summarizes the building specifications of the dormitory building.

Table 1. Building specifications.

Parameter Specific Characteristic Area 172.3 m2 Structure Steel–concrete Envelop Concrete Aboveground Stories 4 Underground Stories 0 Story height 2.6 m Height 10.4 m Width 12.6 m Length 14.5 m Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 5 of 20 consumption standard of a building in Korea was referred from the Korean Energy Agency to build the model. According to this standard, heating, cooling, resident density, lighting density, and equipment density were set as in Table3[ 32]. According to Korean Construction Law, the construction thermal specifications of external wall, internal wall, roof, and floor were set, as listed Table3[ 32–34]. Since this building uses only electricity, cooling and heating were both set from electricity. Table3 briefly summarizes the building specifications of the dormitory building.

Table 2. Climate speciﬁcations of Busan.

Climate Speciﬁcation Value

Latitude 35◦320 N Longitude 129◦190 E Average annual temperature 14.05 ◦C Average max annual temperature 35.4 ◦C Average min annual temperature –8.8 ◦C Average annual relative humidity 63.40% Average solar radiation 106.15 Wh/m2 Average annual wind speed 1.93 m/s Elevation above sea level 33 m

Table 3. Construction speciﬁcations.

Parameter Speciﬁc Characteristic Orientation West to East External wall 0.58 W/m2K Internal wall 2.92 W/m2K Roof 0.27 W/m2K Glazing 3.12 W/m2K Inﬁltration rate 0.3 ac/h Ventilation rate 3 ac/h Equipment 5.38 W/m2 Lighting 3.88 W/m2 Occupancy 0.054 people/m2 Dedicated hot water boiler HVAC/Heating (electricity) HVAC/Cooling Electricity from grid Set point temperature Summer 26 ◦C, Winter 22 ◦C

2.2. Energy and Environmental Analyses According to the building model made using Design Builder in Section 2.1, the one-year energy consumption of the building was predicted. The energy analysis result showed that the one-year energy consumption of the building was 87,830.78 kWh when only electricity was used in the building. On the other hand, when natural gas and electricity were used together, the one-year electricity consumption was 18,013.7 kWh and the one-year natural gas consumption was 69,817.1 kWh. To calculate the annual air pollutants emission of the building, the air pollutant emission densities of electricity and natural gas were referred from the Korea Environmental Industry and Technology Institute, as listed in Table4[ 32,35]. The one-year air pollutants of the building when only electricity was supplied and when natural gas and electricity were supplied together were calculated and compared. The environmental analysis result showed that the amount of CO2 emission was much larger when the building used only electricity than when the building used electricity and natural gas together, as shown in Figure3. On the other hand, emissions of other air pollutants, such as CH 4,N2O, SO2, CO, NOX, and NMVOC, were much lower than CO2 emission. This means that the building using natural gas and electricity together was much cleaner than the building using only electricity for all energy needs. Energies 2019, 12, 3893 6 of 20 Energies 2019, 12, x FOR PEER REVIEW 6 of 20

TableTable 4. AirAir pollutant pollutant emission emission densities densities of of electricity electricity and and natural natural gas.

Air PollutantAir Pollutant Electricity Electricity (kg/kWh) (kg/kWh) Natural Natural Gas (kg/kWh) Gas (kg/kWh) −11 −2 2 CO2 CO2 4.874.87 × 1010− 3.94 × 3.9410 10− × 4 × 4 CH4 CH4 3.503.50 × 1010−−4 2.01 × 2.0110−4 10− × 6 × 7 N2O 1.35 10 1.72 10 N2O 1.35 ×× 10−−6 1.72 × 10−7 × − SO 0.00 100 4.22 10 4 2 0 −4 − SO2 0.00 ×× 10 5 4.22 × 10 × 5 CO 5.00 10− 9.91 10− × −5 −5× NO CO 5.001.20 × 1010 4 9.91 × 5.6710 10 4 X × − × − NMVOCNOX 1.202.00 × 1010−4 5 5.67 × 1.0410−4 10 6 × − × − NMVOC 2.00 × 10−5 1.04 × 10−6 Using only electricity Using natural gas and electricity

42,792.9

CO2CO2

11,527.4

0.0 10,000.0 20,000.0 30,000.0 40,000.0 50,000.0 Amount of air pollutant (kg/year) (a) Using only electricity Using natural gas and electricity 1.8 NMVOC 0.4 10.5 NOXNOx 41.7 4.4 CO 7.8

SO2 0.0 SO2 29.5

N2O 0.1 N2O 0.0 30.7 CH4CH4 20.3 0.010.020.030.040.050.0 (kg/year) Amount of air pollutants (b)

FigureFigure 3. 3. ComparisonComparison of of air air pollutant pollutant amounts amounts between between the the building building using using only only electricity electricity and and the the

buildingbuilding using using electricity electricity and and natural natural gas gas together. together. (a) (aComparison) Comparison of amount of amount of CO of2 CO emission2 emission (b) Comparison(b) Comparison of amount of amount of other of other air pollutants air pollutants emission. emission.

3.3. Modeling Modeling of of SOFC–CGS SOFC–CGS TheThe residential residential fuel fuel cell cell as as a clean energy device is used widely in Japan. Since environmental problemsproblems including including air air pollution pollution have have been been getting getting worse worse recently recently in in Korea, Korea, implementation of of the the residentialresidential fuelfuel cellcell as as a cleana clean energy energy device device in buildings in buildings in Korea in Korea is necessary. is necessary. Therefore, Therefore, this research this researchanalyzed analyzed the eﬃciency the efficiency and environmental and environmenta impact ofl operatingimpact of SOFC–CGSoperating SOFC–CGS in a dormitory in a buildingdormitory in buildingKorea, as in mentioned Korea, as inmentioned previous in sections. previous sections. TypicalTypical approaches of fuel cell cell modeling modeling in in recent recent research research were were summarized summarized in Table Table 55.. In order toto evaluate evaluate and and optimize optimize the the performance performance of of fuel fuel cell, cell, electrochemical electrochemical models models were were developed developed to to simulatesimulate the the specific speciﬁc chemical chemical and and energy energy interaction interaction inside inside the the cell cell stack stack [36–38]. [36–38]. On On the the other other hand, hand, in order to evaluate and improve the device operation efficiency and CO2 reduction by using fuel cell for transportation or residential, fuel cell modeling by mainly using device efficiency was developed

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 7 of 20

in order to evaluate and improve the device operation efficiency and CO2 reduction by using fuel cell for transportation or residential, fuel cell modeling by mainly using device efficiency was developed to simulate operating fuel cell in household or for EV [11,22,24]. In this research, the modeling of SOFC–CGS by mainly using the efficiency of the device was developed to simulate the application of SOFC–CGS in a dormitory building in Korea.

Table 5. Typical approaches of fuel cell modeling.

System System Application Model Application Main Design Variables Volumetric rate generated thermal SOFC [36] Cell stack Electrochemical model energy, rate of production of species, area specific resistance Partial pressure of methane, reaction SOFC–CGS [37] Stationary Electrochemical model rate of steam reforming, cell temperature Partial pressure of species, anodic and SOFC [38] Cell Stack Electrochemical model cathodic charge-transfer coefficients, exchange current density Efficiency of electricity generation, SOFC–CGS [22] Stationary Device efficiency efficiency of heat collection, water temperature Electric demand of cafeteria, required SOFC–CGS [11] Stationary/Transportation Device efficiency electricity for EV charging, electric power generated by SOFC–CGS Power generation efficiency of the fuel SOFC–CHP/PEMFC–CHP [24] Stationary Device efficiency cell, heat recovery efficiency of the fuel cell, heat loss rate of the tank

3.1. Explanation of SOFC–CGS Residential SOFC, as a clean energy supply device, has relatively low efficiency. Due to heat losses, the highest electricity generation efficiency of SOFC is about 46.4% [24,27]. To improve the efficiency of SOFC, SOFC–CGS was developed [24,25,27]. Heat losses from SOFC are collected to heat water, and a backup boiler is used for reheating the water to meet the demand temperature. A 28 L water storage tank was used in this system, as illustrated in Figure4. SOFC uses natural gas to generate electricity to supply the electricity demand of a household. The heat losses of SOFC are collected and transferred to the water storage tank by a water tube. The limit temperature control of the storage tank is 65 ◦C, and the temperature of the storage tank is adjusted by controlling the water flow rate in the tube that collects heat from the SOFC. If the input water temperature through the radiator is higher than 34 ◦C, the radiator reduces the water temperature to 34 ◦C, then the water can enter the SOFC to collect heat. Hot water outputs from the storage tank to the household when hot water is needed. In this system, the city water temperature was set as 15 ◦C, and the hot water temperature demand was set as 40 ◦C. The output water temperature from the water mix device was set to be equal to or less than 33 ◦C. If the water output from the storage tank was equal to or less than 33 ◦C, hot water directly entered the backup boiler for reheating of the water to 40 ◦C and then the hot water at 40 ◦C was provided to the household. If the output water temperature from the storage tank was higher than 33 ◦C, the water mix device mixed city water and water from the storage tank to attain the temperature of 33 ◦C, and then, through the backup boiler, water was heated to 40 ◦C to supply the household. This system is illustrated as Figure4. The specifications of the SOFC–CGS in this study are shown in Table6, and the efficiency of electricity generation and heat collection of the SOFC–CGS are shown in Figure5. The mathematical equation of the electricity generation efficiency is written as Equation (1). The efficiency of heat collection of the SOFC–CGS was constantly 31%.

EFF = 0.6868LF3 1.6829LF2 + 1.4601LF. (1) ElSup − Energies 2019, 12, 3893 8 of 20 Energies 2019, 12, x FOR PEER REVIEW 8 of 20

Electricity supply

Hot water supply Backup SOFC Water storage boiler tank Household

28 L

Energies 2019, 12, x FOR PEER REVIEWRadiator 8 of 20 Natural gas City water Natural gas

Figure 4. WorkingWorking principle of the Electricity solid oxide supply fuel cell cogeneration system (SOFC–CGS).

60% Table 6. Speciﬁcations of the SOFC–CGS. Efficiency of electricity generation Efficiency of heat collection Hot water supply Type of Cell StackBackup SOFC (Solid Oxide Fuel Cell) SOFC Water 50% boiler Rated electrical outputstorage 700 W Max running timetank Household26 days Standby40% time for start28 L 1 day Water tank volume 28 L Reaction to overheating Heat dissipation by radiator Eﬃciency of30%Radiator heat collection 31%

Water temperatureNatural gas fromEfficiency tank to backupCity water boiler Natural gas 40 ◦C Eﬃciency of electricity generation 46.4% Figure 4. Working 20%principle of the solid oxide fuel cell cogeneration system (SOFC–CGS).

60% 10% Efficiency of electricity generation Efficiency of heat collection 50% 0% 0% 20% 40% 60% 80% 100% 40% Load factor of electricity generation Figure 5. Efficiency of the SOFC–CGS. 30%

Efficiency Table 6. Specifications of the SOFC–CGS.

Type20% of Cell Stack SOFC (Solid Oxide Fuel Cell) Rated electrical output 700 W Max10% running time 26 days Standby time for start 1 day

Water0% tank volume 28 L Reaction 0%to overheating 20% 40% 60%Heat 80% dissipation 100% by radiator Efficiency of heat collectionLoad factor of electricity generation 31% Water temperature from tank to backup boiler 40 °C Figure 5. EfficiencyEﬃciency of the SOFC–CGS. Efficiency of electricity generation 46.4%

In the equation, EFFElSup isTable the e ﬃ6. ciencySpecifications of electricity of the SOFC–CGS. generation of the SOFC and LF is the load 3.2.factor Mathemetical of electricity Model generation. of SOFC–CGS Type of Cell Stack SOFC (Solid Oxide Fuel Cell) In this study, the mathematical model of SOFC–CGS was referred from recent research from 3.2. Mathemetical ModelRated of SOFC–CGS electrical output 700 W Kyushu University [25–27].Max Therunning electricity time generation calculation follows26 days the steps below. Firstly, dependingIn this on study, the electricity theStandby mathematical generation time for model starttracking of SOFC–CGS speed, the was maximum referred electricity1 fromday recent supply research capability from canKyushu be determined University by [25 Equation–Water27]. The tank (2): electricity volume generation calculation follows28 L the steps below. Firstly,

Reaction to overheatingPSupmax = − PElSup + VElSupSpIHeattint, dissipation by radiator (2) Efficiency of heat collection 31% where PSupmaxWater is temperature the maximum from electricity tank to backup supply boiler capability, PElSup is40 the °C amount of electricity generation, VElSupSpEfficiency is the electricity of electricity generation generation tracking speed, and Itint is the46.4% calculation time interval. Then, according to the household electricity demand, the electricity supply from SOFC can be

3.2.Energies Mathemetical 2019, 12, x; doi: Model FOR ofPEER SOFC–CGS REVIEW www.mdpi.com/journal/energies In this study, the mathematical model of SOFC–CGS was referred from recent research from Kyushu University [25–27]. The electricity generation calculation follows the steps below. Firstly, depending on the electricity generation tracking speed, the maximum electricity supply capability can be determined by Equation (2):

PSupmax = −PElSup + VElSupSpItint, (2) where PSupmax is the maximum electricity supply capability, PElSup is the amount of electricity generation, VElSupSp is the electricity generation tracking speed, and Itint is the calculation time interval. Then, according to the household electricity demand, the electricity supply from SOFC can be

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 9 of 20 depending on the electricity generation tracking speed, the maximum electricity supply capability can be determined by Equation (2):

P = P + V I , (2) Supmax − ElSup ElSupSp tint where PSupmax is the maximum electricity supply capability, PElSup is the amount of electricity generation, VElSupSp is the electricity generation tracking speed, and Itint is the calculation time interval. Then, according to the household electricity demand, the electricity supply from SOFC can be determined. Following Equations (3) and (4), the amount of natural gas consumption by SOFC can be calculated:

PElSup LF = (3) PElMax

PElSup PEhCell = , (4) EFFElSup where LF is the load factor of electricity generation, PElMax is maximum electricity generation capacity of SOFC, PEhCell is the amount of natural gas consumption of SOFC, and EFFElSup is the eﬃciency of electricity generation. The amount of heat output from SOFC can be calculated by Equation (5):

PHtSup = PEhCellEFFHtSup, (5) where PHtSup is the amount of heat output from SOFC and EFFHtSup is the heat output efficiency of SOFC, which is 31%. The water flow rate in the water tube that collects heat can be calculated using either Equation (6) or Equation (7). If the input water temperature through the radiator unit is higher than or equal to 34 ◦C, Equation (6) is used to calculate the water flow rate in water tube; otherwise, Equation (7) is used to calculate it. PHtSup VwCell = (6) Cw(T 34) out − PHtSup VwCell = . (7) Cw(T T ) out− in In the equations, VwCell is the water flow rate in the water tube; PHtSup is the amount of heat supply from SOFC; Tout is the target output water temperature from the SOFC, which was set as 65 ◦C in this study; Tin is the input water temperature to the radiator unit; and Cw is the specific heat of water. If the water temperature is higher than 34 ◦C, the heat dissipation of the radiator unit is determined by Equation (8); otherwise, the heat dissipation of the radiator unit is zero:

P = V Cw(Tout T ), (8) Rad wCell − in where PRad is the heat dissipation from the radiator unit. The amount of heat output from the backup boiler is determined by Equation (9), and the amount of gas used in the backup boiler is calculated from Equation (10): P = Vol Cw(T T ), (9) BBSup HwDmd Dmd− BBin PBBSup PBBUse = , (10) EFFBB where PBBSup is the amount of heat output from the backup boiler, PBBUse is the amount of gas used in the backup boiler, TDmd is the water temperature demand from the household, TBBin is the input water temperature to the backup boiler, VolHwDmd is the amount of hot water demand from the household, PBBUse is the amount of gas used in the backup boiler; and EFFBB is the working eﬃciency of the backup boiler, which was set as 95% in this study. Energies 2019, 12, 3893 10 of 20

The calculation ﬂow of the program and the input data are shown as Figure6 and Table7. Energies 2019, 12, x FOR PEER REVIEW 10 of 20

Start

Specifications setting of SOFC-CGS

Input data of time, weather, city water and electricity, hot water demand from household

Calculate electricity generation, heat output and gas use of SOFC

Calculate water flow rate in water tube of heat collection

Calculate water temperature change in water storage tank No

If input water temperature to No radiator is above 34 ℃

Yes Radiator stand by Calculate heat dissipation of radiator

Calculate gas consumption of backup boiler

Store calculation results

Input data finished by end time

Yes End

FigureFigure 6. 6. CalculationCalculation diagram diagram of of the the program. program.

TableTable 7. 7. InputInput data data of of the the program. program.

InputInput Data Data Unit/Value Unit/Value ElectricityElectricity demand demand W W Hot waterHot water demand demand L/Time L/Time unit unit SettingSetting temperature temperature of backup of backup boiler boiler 40 °C 40 ◦C Outside temperature Data from ASHRAE Outside temperature Data from ASHRAE Temperature of city water 15 ◦C Temperature of city water 15 °C Temperature of hot water demand 40 ◦C Temperature of hot water demand 40 °C 4. Simulation of Operating SOFC–CGS in the Dormitory Building 4. Simulation of Operating SOFC–CGS in the Dormitory Building As mentioned in Section2, the energy consumption for heating the research target building in As mentioned in Section 2, the energy consumption for heating the research target building in winter was significant. Since using electricity from the public power plant results in much more air winter was significant. Since using electricity from the public power plant results in much more air pollution than using natural gas for heating and because the efficiency of directly using natural gas for heating was much higher than that of using natural gas to generate electricity for heating by the SOFC, in this research, energy for heating was not considered as being supplied from the SOFC. Aside from heating, the SOFC–CGS provides electricity and hot water to the households. Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 11 of 20 pollution than using natural gas for heating and because the efficiency of directly using natural gas for heating was much higher than that of using natural gas to generate electricity for heating by the SOFC, in this research, energy for heating was not considered as being supplied from the SOFC. Aside from heating,Energies 2019 the, 12 SOFC–CGS, x FOR PEER providesREVIEW electricity and hot water to the households. 11 of 20 The research target building had four floors, and each floor had the same structure with five households.The research As the target energy building consumption had four of each floors, household and each was floor predicted had the by thesame calibrated structure commercial with five softwarehouseholds. Design As Builder,the energy the energyconsumption consumption of each pattern household and amount was predicted were the sameby the for eachcalibrated floor. Thus,commercial the first software floor was Design chosen Builder, for this the research. energy Asconsumption mentioned pattern previously, and thereamount were were five the households same for ineach first floor. floor, Thus, and the first energy floor consumption was chosen patternfor this research. was similar As amongmentioned the fivepreviously, households there in were the firstfive floor.households According in first to floor, schedule and the of ASHRAE energy consumpt standardion for pattern residential was similar house [among31], the the electricity five households and hot waterin the consumptionfirst floor. According patterns to of schedule representative of ASHRAE dates ofstandard each season for residential of household house 1 in [31], the the first electricity floor are shownand hot as water Figure consumption7. As heating patterns energy of was representative provided directly dates of by each natural season gas, of it household was not considered 1 in the first in simulatingfloor are shown the operating as Figure SOFC–CGS. 7. As heating The energy yearly energywas provided consumption directly of eachby natural household gas, it in was the firstnot floorconsidered is shown in assimulating Figure8. the operating SOFC–CGS. The yearly energy consumption of each household in the first floor is shown as Figure 8.

(a)

(b)

(c)

Figure 7.7. Representative daily data of electricity demanddemand and hot water demand of household 1 in each season. ( a) Data in summer; (b) Data in winter; (c) Data in intermediateintermediate seasons.seasons.

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, x FOR PEER REVIEW 12 of 20

Natural gas for heating Electricity Energy for hot water

Household 5 3.04 0.72 0.66

Household 4 3.30 0.64 0.60

Energies 2019, 12, 3893 Household 3 3.70 1.32 0.72 12 of 20 Energies 2019, 12, x FOR PEER REVIEW 12 of 20 First floor Household 2 1.79 0.85 0.60 Natural gas for heating Electricity Energy for hot water

HouseholdHousehold 1 5 2.413.04 0.97 0.720.64 0.66

Household 40.00 1.003.30 2.00 3.000.64 4.00 0.60 5.00 6.00 7.00 Amount of energy (MWh/household*year)

FigureHousehold 8. Yearly 3 energy consumption3.70 of each household1.32 in0.72 the first floor. First floor To improve and analyzeHousehold the 2 efficiency1.79 of SOFC0.85 –CGS,0.60 one SOFC–CGS was simulated to operate in either only one household or in multi-households (two or three households). According to the yearly energy consumptionHousehold of each 1 household,2.41 as shown0.97 in0.64 Figure 8, household 2 ranked the lowest and household 3 ranked the highest. Therefore, in order to form a balance, households 2 and 3 were simulated as sharing one SOFC–CGS,0.00 1.00and households 2.00 3.00 1, 4, 4.00 and 5 5.00 were 6.00simulated 7.00 as sharing one Amount of energy (MWh/household*year) SOFC–CGS. Hourly data of electricity demand, electricity generation, and heat output from the SOFC and the heat reductionFigure amount 8. Yearly from energy the consumption radiator at of representative each household days in the in firstfirst the floor.floor. year in summer are shown as Figures 9–11 below. To improveimprove andand analyzeanalyze thethe eefficiencyfficiency ofof SOFC–CGS,SOFC–CGS, oneone SOFC–CGSSOFC–CGS waswas simulatedsimulated to operate As the data of the simulation results were large, the simulation results of households 2 and 3 inin eithereither onlyonly oneone householdhousehold oror inin multi-householdsmulti-households (two(two oror threethree households).households). According to thethe were selected for presentation in this research. Figures 9 and 10 show the simulation results when yearlyyearly energyenergy consumptionconsumption ofof eacheach household,household, as shown in Figure8 8,, householdhousehold 22 rankedranked thethe lowestlowest operating one SOFC–CGS for one household. Figure 11 shows the simulation results when sharing and household 3 ranked the highest. Therefore, in order to form a balance, households 2 and 3 were one SOFC–CGS by multi-households. The load factor of SOFC–CGS was increased when multi- simulated as sharing one SOFC–CGS, and households 1, 4, and 5 werewere simulatedsimulated as sharingsharing oneone households share one SOFC–CGS, as in Figure 11. This means that the electricity generation efficiency SOFC–CGS. HourlyHourly datadata ofof electricity demand, electricityelectricity generation, andand heatheat output from the SOFC of SOFC–CGS was improved. Additionally, the ratio of heat reduction amount by the radiator was and the heatheat reductionreduction amount from the radiator at representative days in the year in summersummer are reduced when SOFC–CGS was shared by multi-households, as shown in Figure 11, which means that shown asas FiguresFigures9 9–11–11 below. below. the efficiency of heat usage for hot water was increased. As the data of the simulation results were large, the simulation results of households 2 and 3 were selected for presentation(w) Electricityin this research. generation fromFigures SOFC 9 and Electricity10 show demand the simulation results when operating one SOFC–CGS1200 for one household. Figure 11 shows the simulation results when sharing one SOFC–CGS by1000 multi-households. The load factor of SOFC–CGS was increased when multi- households share one800 SOFC–CGS, as in Figure 11. This means that the electricity generation efficiency of SOFC–CGS was600 improved. Additionally, the ratio of heat reduction amount by the radiator was reduced when SOFC–CGS400 was shared by multi-households, as shown in Figure 11, which means that

the efficiency of heatenergy of Amount usage for hot water was increased. 200

(w0 ) Electricity generation from SOFC Electricity demand 1200 (a) (w) Radiator energy reduction Heat supply from SOFC 5001000 450800 400 350600 300 250400

Amount of energy of Amount 200 150200

Amount of energy of Amount 100 500 0 (a) (w) 0120120Radiator energy reduction Heat supply from SOFC 500 450 7/30 7/31 400 350 (b) 300 Figure 9. Hourly250 simulation data of household 2. (a) Data of electricity generation and electricity 200 demand; (b) Data150 of heat reduction and heat supply. Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies

Amount of energy of Amount 100 50 0 0120120

7/30 7/31 (b)

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019,, 12,, xx FORFOR PEERPEER REVIEWREVIEW 13 of 20

EnergiesFigure2019, 129., Hourly 3893 simulation data of household 2. (a) Data of electricity generation and electricity13 of 20 demand; (b) Data of heat reduction and heat supply.

(w) (w) Electricity generation from SOFC Electricity demand 2000

1500

1000 Amount of energy of Amount Amount of energy of Amount 500

0 (a) (w)(w) Radiator energy reduction Heat supply from SOFC 500 450 400 350 300 250 200 150

Amount of energy of Amount 100 Amount of energy of Amount 100 50 0 0120120 7/30 7/31 (b)

Figure 10. Hourly simulation data of household 3. ( (aa)) Data Data of of electricity generation and electricity demand; ( b) Data of heat reduction andand heatheat supply.supply.

(w) (w) Electricity generation from SOFC Electricity demand 3000 2500 2000 1500 1000 Amount of energy of Amount Amount of energy of Amount 500 0 (a) (w)(w) Radiator energy reduction Heat supply from SOFC 500 450 400 350 300 250 200 150

Amount of energy of Amount 100 Amount of energy of Amount 100 50 0 0120120 7/30 7/31 7/30 7/31 (b)

Figure 11. HourlyHourly simulation simulation data data of of households 2 and 3. ( (aa)) Data Data of of electricity generation and electricity demand; ( b) Data of heat reductionreduction andand heatheat supply.supply.

As the data of the simulation results were large, the simulation results of households 2 and 3 were selected for presentation in this research. Figures9 and 10 show the simulation results when operating one SOFC–CGS for one household. Figure 11 shows the simulation results when sharing one SOFC–CGS by multi-households. The load factor of SOFC–CGS was increased when multi-households Energies 2019,, 12,, x;x; doi:doi: FORFOR PEERPEER REVIEWREVIEW www.mdpi.com/journal/energies Energies 2019, 12, x FOR PEER REVIEW 14 of 20

5. Analysis of Efficiency and Environmental Impact of SOFC–CGS

5.1. Analysis of SOFC–CGS Efficiency Energies 2019, 12, 3893 14 of 20 The yearly energy consumptions of simulating the operation of one SOFC–CGS to one household and multi-households are shown as Figure 12. The natural gas consumption by SOFC, electricityshare one SOFC–CGS,generation asby in SOFC, Figure purchased 11. This means electricity, that the and electricity natural generation gas consumption efficiency by of the SOFC–CGS backup wasboiler improved. were higher Additionally, when simulating the ratio the ofoperation heat reduction of one SOFC–CGS amount by for the multi-households radiator was reduced than when simulatingSOFC–CGS for was one shared household. by multi-households, The electricity ascons shownumption in Figure by SOFC 11, which was cons meanstant that because the effi electricityciency of washeat only usage used for hotwhen water the SOFC was increased. started or restarted. Increasing electricity generation led to increasing efficiency of electricity generation by SOFC. The reason for this was that sharing one SOFC–CGS to 5. Analysis of Efficiency and Environmental Impact of SOFC–CGS multi-households led to increasing electricity and hot water demand for the one SOFC–CGS. 5.1. AnalysisAccording of SOFC–CGS to Figure 13, Effi theciency electricity generation efficiency of SOFC ranged from 13% to 22% and the heat usage efficiency ranged from 22% to 23% when simulating the operation of one SOFC– The yearly energy consumptions of simulating the operation of one SOFC–CGS to one household CGS for one household. The efficiency of electricity generation of SOFC was increased to range from and multi-households are shown as Figure 12. The natural gas consumption by SOFC, electricity 28% to 29% and the heat usage efficiency was increased to range from 24% to 25% when multi- generation by SOFC, purchased electricity, and natural gas consumption by the backup boiler were households share one SOFC–CGS. The percentages of losses from different components of SOFC– higher when simulating the operation of one SOFC–CGS for multi-households than when simulating CGS of simulating operation of one SOFC–CGS to one household and sharing one SOFC–CGS by for one household. The electricity consumption by SOFC was constant because electricity was only multi-households are compared in Figure 14. The losses in fuel cell and radiator were decreased when used when the SOFC started or restarted. Increasing electricity generation led to increasing efficiency of multi-households share one SOFC–CGS. In another aspect, sharing one SOFC–CGS by multi- electricity generation by SOFC. The reason for this was that sharing one SOFC–CGS to multi-households households increased the electricity demand from the public power plant, which leads to more air led to increasing electricity and hot water demand for the one SOFC–CGS. pollution. The specific environmental analysis is shown in Section 5.2.

Gas consumption of FC Electricity generation Purchased electricity Gas consumption of backup boiler Electricity consumption of FC

Households 1, 4, and 5 6.4 1.8 0.5

Households 2 and 3 6.3 1.8 0.4

Household 5 4.7 0.70.1

Household 4 4.6 0.60.1

Household 3 5.5 1.2 0.1

Household 2 4.9 0.8 0.0 Use 1 FC to 1 householdto FC 1 Use FC Share Household 1 5.0 0.9 0.1

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Amount of energy (MWh/household*year) Figure 12. 12. YearlyYearly energy energy consumptions consumptions of simulating of simulating one SOFC–CGS one SOFC–CGS for one household for one householdand multi- households.and multi-households.

According to Figure 13, the electricity generation efficiency of SOFC ranged from 13% to 22% and the heat usage efficiency ranged from 22% to 23% when simulating the operation of one SOFC–CGS for one household. The efficiency of electricity generation of SOFC was increased to range from 28% to 29% and the heat usage efficiency was increased to range from 24% to 25% when multi-households share one SOFC–CGS. The percentages of losses from different components of SOFC–CGS of simulating operation of one SOFC–CGS to one household and sharing one SOFC–CGS by multi-households are compared in Figure 14. The losses in fuel cell and radiator were decreased when multi-households share one SOFC–CGS. In another aspect, sharing one SOFC–CGS by multi-households increased the electricity demand from the public power plant, which leads to more air pollution. The specific environmental analysis is shown in Section 5.2. Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 15 of 20 Energies 2019, 12, x FOR PEER REVIEW 15 of 20

Energies 2019, 12, x FOREfficiency PEER REVIEW of heat usage Electricity efficiency generation of FC 15 of 20 Efficiency of heat usage Electricity efficiency generation25% of FC Households 1, 4, and 5 29% Households 1, 4, and 5 25% 24% 29% Households 2 and 3 28% 24% Households 2 and 3 28% Household 5 22% 14% 22% Household 5 14% Household 4 22% 13% 22% Household 4 13% Household 3 23% 23%22% Household 3 22% Household 2 22% 17% 22% Household 2 17% Use Use 1 FC to 1 household Share FC Household 1 22% Use Use 1 FC to 1 household Share FC 18% 22% Household 1 18% 0% 5% 10% 15% 20% 25% 30% 35% 0% 5% 10% 15% 20% 25% 30% 35% Efficiency Efficiency Figure 13. Yearly electricity generation and heat output efficiencies of simulating using one SOFC FigureFigure 13. 13.Yearly Yearly electricity electricity generation generation andand heat output efficiencies eﬃciencies of of simulating simulating using using one one SOFC SOFC cogeneration system for one household and multi-households. cogenerationcogeneration system system for for one one household household andand multi-households.multi-households.

ElectricityElectricity generation generation HeatHeat usage(efficiency) usage(efficiency) Loss inin radiatorradiator LossLoss in in storage storage tank tank LossLoss in FCin FC

HouseholdsHouseholds 1, 1,4, 4, and and 5 5 29%29% 25% 5%5%2%2% 40%40%

HouseholdsHouseholds 2 and2 and 3 3 28%28% 24% 5%5%2%2% 41%41%

HouseholdHousehold 5 5 14%14% 22%22% 6% 2%2% 55%55%

HouseholdHousehold 4 4 13%13% 22%22% 6% 2%2% 56%56%

HouseholdHousehold 3 3 22%22% 23% 6%6%2%2% 47%47%

Household 2 17% 22% 7% 2% 53% Household 2 17% 22% 7% 2% 53% Use 1 FC to 1 householdFC to 1 Use FC Share Use 1 FC to 1 householdFC to 1 Use FC Share Household 1 18% 22% 7% 2% 51% Household 1 18% 22% 7% 2% 51%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20%Percentage 30% of 40% energy usage 50% and 60% losses 70% 80% 90% 100% Percentage of energy usage and losses FigureFigure 14. 14. YearlyYearly energy energy losses losses of simulating of simulating using one using SOFC–CGS one SOFC–CGS to one household to one and household multi- Figure 14. Yearly energy losses of simulating using one SOFC–CGS to one household and multi- andhouseholds. multi-households. households. 5.2.5.2. Analysis Analysis of of Environmental Environmental Impact Impact ofof SOFC–CGSSOFC–CGS 5.2. Analysis of Environmental Impact of SOFC–CGS IncludingIncluding the the two two previous previous casescases ofof energyenergy supply types types in in the the building building in in Section Section 2,2 ,four four diffdifferenterentIncluding cases cases ofthe of energy energytwo supplyprevious supply types types cases are are of explainedexplained energy supply andand summarized types in inthe in Table Tablebuilding 8.8 .In In the thein firstSection first case, case, 2,only onlyfour electricitydifferentelectricity cases was was of supplied supplied energy tosupply to the the building, building, types are whilewhile explained in thethe an second d summarized case, case, natural natural in Tablegas gas and and8. Inelectricity electricity the first together case, together only wereelectricitywere supplied. supplied. was supplied According According to the to to the thebuilding, analysisanalysis while result,result, in th thee second energycase, energy natural supply supply gas type type and emitted emitted electricity a relatively a relatively together smallerweresmaller supplied. amount amount According of of air air pollutants pollutants to the than analysisthan thethe firstfirstresult, type the and second the the efficiency e ffienergyciency supplyof of using using type natural natural emitted gas gas directly a directly relatively by by combustionsmallercombustion amount was was of much airmuch pollutants higher higher than than than usingusing the first naturalnatural type gasgaands by the SOFC efficiency for for heating. heating. of using Therefore, Therefore,natural gas in in thedirectly the third third by case,combustioncase, one one SOFC–CGS SOFC–CGS was much waswas higher applied than to to using one one household householdnatural ga unders under by theSOFC the second secondfor heating.energy energy supply Therefore, supply type, type,andin the in and thethird in fourth case, one SOFC–CGS was applied to multi-households under the second energy supply type. thecase, fourth one SOFC–CGS case, one SOFC–CGS was applied was to one applied household to multi-households under the second under energy the supply second type, energy and supply in the fourthThe case,environmental one SOFC–CGS analysis was for appliedthe entire to buildingmulti-households (four stories) under under the thesecond four energy different supply cases type.of The environmental analysis for the entire building (four stories) under the four different cases of Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 16 of 20 type. The environmental analysis for the entire building (four stories) under the four different cases of Energies 2019, 12, x FOR PEER REVIEW 16 of 20 energy application types is shown as Figure 15. According to the result of the environmental analysis, theenergy amount application of CO2 emissiontypes is shown was most as Figure significant 15. Acco in therding case to the of using result only of the electricity environmental in the building.analysis, Thethe amount CO2 emission of CO2 of emission the second was case, most using significant natural in gasthe andcase electricityof using only together electricity in the in building, the building. was muchThe CO lower2 emission than in of the the case second of using case,only using electricity natural ga ins theand building. electricity Under together the energyin the building, supply type was ofmuch the secondlower than case, in when the case SOFC–CGS of using wasonly simulated electricityas in operatingthe building. in the Under building, the energy the CO supply2 emission type wasof the significantly second case, reduce when to SOFC–CGS range from 3057.2was simulated to 4120.6 as kg operating/year. The in case the of building, operating the one CO SOFC–CGS2 emission forwas one significantly household reduce gave theto range lowest from CO 3057.22 emission. to 4120.6 Due kg/year. to the purchased The case of electricity operating fromone SOFC–CGS the public powerfor one plant household was increased gave the from lowest the CO third2 emission. case to the Due fourth to the case purchase as shownd electricity in the energy from analysis the public in Figurepower 12 plant, the was fourth increased case resulted from inthe a third higher case air pollutantsto the fourth emission case as than shown the in third the case. energy However, analysis the in caseFigure of sharing12, the fourth one SOFC–CGS case resulted among in a multi-households higher air pollutants had aemission higher system than the effi thirdciency case. than However, the case ofthe operating case of sharing one SOFC–CGS one SOFC–CGS for one among household, multi-hous as indicatedeholds byhad the a higher analysis syst resultem efficiency in Section than 5.1. the In anothercase of operating aspect, the one diff SOFC–CGSerence of CO for2 emissionone household, between as casesindicated three by and the four analysis was notresult significant, in Section as 5.1. in FigureIn another 15; therefore, aspect, the the difference case of sharing of CO2 oneemission SOFC–CGS between among cases multi-households three and four was was not recommended. significant, as Accordingin Figure to15; Figure therefore, 15, the the emission case amountsof sharing of theone pollutants SOFC–CGS other among than CO multi-households2 of the four different was casesrecommended. were much According lower than to the Figure CO2 15,emissions the emissi ofon the amounts four diff erentof the cases. pollutants other than CO2 of the four different cases were much lower than the CO2 emissions of the four different cases. Table 8. Explanation of cases for environmental analysis.

Table 8. Explanation of cases for environmental analysis.Energy for Other Needs in Case Number Case Name Energy for Heating/Hot Water Household Case Energy for Case Name Purchased electricity fromEnergy power for PurchasedOther Needs electricity in Household from power Number1 Using only electricityHeating/Hot Water plant plant Using only Purchased electricity 1 Using natural gas and Direct natural gas (heatingPurchased and electricityPurchased electricityfrom power from plant power 2 electricityelectricity from power planthot water) plant Using natural gas Direct natural gas Individual SOFC–CGS for one 2 One SOFC–CGS for one Purchased electricity from power plant 3 and electricity (heating andDirect hot natural water) gas (heating) household and purchased electricity household One SOFC–CGS for Direct natural gas Individual SOFC–CGSfrom for power one planthousehold 3 one household (heating) and purchased electricityShared SOFC–CGSfrom power for plant One SOFC–CGS to 4 One SOFC–CGS to Direct naturalDirect natural gas gas (heating)Shared SOFC–CGSmulti-households for multi-households and purchased and 4 multi-households multi-households (heating) purchased electricityelectricity from from power power plant

(a)

Figure 15. Cont.

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 17 of 20 Energies 2019, 12, x FOR PEER REVIEW 17 of 20

One SOFC-CGS to multi-households One SOFC-CGS to one household Using only electricity Using natural gas and electricity 0.1 0.1 NMVOC 1.8 0.4 34.9 34.5 NOXNOx 10.5 41.7 6.2 6.1 CO 4.4 7.8 25.7 SO2 25.6 SO2 0.0 29.5 0.0 0.0 N2ON2O 0.1 0.0 13.5 12.7 CHCH44 30.7 20.3 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 (kg/year) Amount of air pollutants (b)

Figure 15.15. ComparisonComparison of of air air pollutant amounts of the building under the four didifferentﬀerent energy supplysupply types.types. ( (aa)) Comparison of of amount of CO2 emission ( b)) Comparison of amount of other airair pollutantspollutants emission.emission.

6. Conclusions Residential SOFC,SOFC, asas a clean energy device, continuescontinues to be researched by researchers in the fieldfield of architecture.architecture. However, as the eefficiencyfficiency of SOFCSOFC is relatively low, SOFC–CGSSOFC–CGS hashas beenbeen studiedstudied recentlyrecently byby manymany researchers.researchers. In this research, thethe environmentalenvironmental and eefficiencyfficiency impactsimpacts ofof operatingoperating SOFC–CGS in aa dormitorydormitory buildingbuilding inin KoreaKorea werewere analyzed.analyzed. As aa newlynewly developeddeveloped building,building, thethe researchresearch targettarget buildingbuilding waswas constructedconstructed withwith electricityelectricity asas thethe onlyonly energyenergy resource.resource. AnalysisAnalysis of thisthis strategy was was required. required. Firstly, Firstly, in in this this study, study, the the environmental environmental impact impact was wascompared compared between between the cases the casesof using of using only onlyelectricity electricity (new (new way) way) and and using using elec electricitytricity and and natural natural gas gas (traditional (traditional way) way) in thethe building. The analysis result show showeded that the traditional way of energyenergy supply type in thethe buildingbuilding emits less airair pollutionpollution than the newnew way.way. Secondly, based on thethe traditionaltraditional energy supply type, operating SOFC–CGS in the buildingbuilding waswas simulated.simulated. The eefficiencyfficiency andand environmentalenvironmental impact werewere analyzed, and these were compared between the casescases of simulating one SOFC–CGS in one household and sharing one SOFC–CGS among among multi-households multi-households.. The The specific specific conclusions conclusions are are listed listed below. below. (1)(1) TheThe comparisoncomparison analysisanalysis betweenbetween supplyingsupplying only electricityelectricity to thethe buildingbuilding andand supplyingsupplying electricity and natural gas together showed that supplyingsupplying electricity and natural gas together results inin muchmuch lessless airair pollutionpollution thanthan supplyingsupplying onlyonly electricity.electricity. (2)(2) UnderUnder the the energy energy supply supply type type of supplying of supplying electricity electricity and natural and natural gas together gas together to the building, to the SOFC–CGSbuilding, SOFC–CGS was simulated was simulated in the building in the under building the conditionsunder the conditions of operating of one operating SOFC–CGS one SOFC– in one householdCGS in one and household sharing one and SOFC–CGS sharing amongone SOFC–C multi-households.GS among Themulti-households. electricity generation The electricity efficiency ofgeneration SOFC ranged efficiency from of 13% SOFC to ranged 22% and from the 13% heat to usage 22% eandfficiency the heat ranged usage from efficiency 22% to ranged 23% when from simulating22% to 23% operation when simulating of one SOFC–CGS operation in of one one household. SOFC–CGS The in electricity one household. generation The effi electricityciency of SOFCgeneration increased efficiency to range of fromSOFC 28% increased to 29% andto rang the heate from usage 28% effi tociency 29% increasedand the heat to range usage from efficiency 24% to 25%increased when to multi-households range from 24% sharedto 25% onewhen SOFC–CGS. multi-households shared one SOFC–CGS. (3)(3) TheThe fourfour didifferentfferent cases cases of of energy energy supply supply type type for for environmental environmental analysis analysis were were summarized summarized in Table8. According to the results of the environmental analysis, the CO emission was significantly in Table 8. According to the results of the environmental analysis, the CO22 emission was significantly reduced from 11,527.4 kg/year kg/year to to 3057.2 3057.2 kg/year kg/year when when simulating simulating the the operation operation of ofone one SOFC–CGS SOFC–CGS in in one household under the energy supply type of case two. On the other hand, the CO emission one household under the energy supply type of case two. On the other hand, the CO2 emission2 was significantly reduced from 11,527.4 kg/year to 4120.6 kg/year when simulating the sharing of one

Energies 2019, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/energies Energies 2019, 12, 3893 18 of 20 was significantly reduced from 11,527.4 kg/year to 4120.6 kg/year when simulating the sharing of one SOFC–CGS among multi-households. The emission amounts of pollutants other than CO2 of the four different cases were much lower than the CO2 emissions of the four different cases. (4) The difference of CO2 emission between a simulating operation of one SOFC–CGS in one household and the sharing of one SOFC–CGS among multi-households was not significant. Meanwhile, simulating the sharing one SOFC–CGS among multi-households led to higher efficiency than simulating the operation of one SOFC–CGS in one household. Therefore, sharing one SOFC–CGS among multi-households in a dormitory building is recommended.

Author Contributions: H.C. conceived and completed this paper. I.-H.L. supervised and reviewed this paper. Funding: This work was supported by a National Research Foundation of Korea grant funded by the Korean Government (NRF-2018R1D1A1B07051106). Conﬂicts of Interest: The authors declare no conﬂict of interest.

Abbreviations

SOFC Solid oxide fuel cell SOFC–CGS Solid oxide fuel cell cogeneration system LNG Liquefied Natural Gas CHP Combined heat and power DC Direct current AC Alternating current EV Electric vehicle SOFC–CHP Solid oxide fuel cell combined heat and power GHG Greenhouse gas HVAC Heating, Ventilation and Air Conditioning ASHRAE American Society of Heating, Refrigerating and Air-Conditioning Engineers PEMFC–CHP Proton exchange membrane fuel cell combined heat and power Nomenclatures EFFElSup Efficiency of electricity generation LF Load factor of electricity generation PSupmax Maximum electricity supply capability PElSup Amount of electricity supplication VElSupSp Electricity generation tracking speed Itint Time interval PElMax Maximum electricity generation capacity of SOFC PEhCell Amount of natural gas consumption of SOFC PHtSup Amount of heat output from SOFC EFFHtSup Heat output efficiency of SOFC VwCell Water flow rate in the water tube Tout Target output water temperature from the SOFC Tin Input water temperature to the radiator unit Cw Specific heat of water PRad Heat dissipation from the radiator unit PBBSup Amount of heat output from the backup boiler PBBUse Amount of gas used in the backup boiler TDmd Water temperature demand from the household TBBin Input water temperature to the backup boiler VolHwDmd Amount of hot water demand EFFBB Working efficiency of the backup boiler

References

1. Doosoon, L. An Analysis of the Eﬀect of Relative Prices between Electricity and Gas on the Sectoral Demand of City Gas. Master’s Thesis, Public Administration of Seoul National University, Seoul, Korea, 2014. Energies 2019, 12, 3893 19 of 20

2. Korea Electric Power Corporation. Sustainable Report. 2016. Available online: www.kepco.co.kr (accessed on 6 August 2018). 3. Wang, Y.; Zhao, Q.; Borup, U.; Choi, S. Stack and converter-integrated design for efficient residential fuel cell system. Int. J. Hydrogen Energy 2009, 34, 7316–7322. [CrossRef] 4. Yuqing, W.; Yixiang, S.; Meng, N.; Ningsheng, C. A micro tri-generation system based on direct flame fuel cells for residential applications. Int. J. Hydrogen Energy 2014, 39, 5996–6005. 5. Hirohisa, A.; Tetsuya, W.; Ryohei, Y. Development of an energy management system for optimal operation of fuel cell based residential energy systems. Int. J. Hydrogen Energy 2016, 41, 20314–20325. 6. Alexandros, A.; Eric, S.F.; Dan, J.L.B. Options for residential building services design using fuel cell based micro-CHP and the potential for heat integration. Appl. Energy 2015, 138, 685–694. 7. Mohammad, H.-M. Transient performance improvement of a stand-alone fuel cell for residential applications. Sustain. Cities Soc. 2019, 50, 101650. 8. Hong, B.; Qiong, W.; Weijun, G.; Weisheng, Z. Optimal operation of a grid-connected hybrid PV/fuel cell/battery energy system for residential applications. Energy 2016, 113, 702–712. 9. Yasuhiro, H.; Kiyotaka, T.; Ryuichirao, G.; Hideki, K. Hybrid utilization of renewable energy and fuel cells for residential energy systems. Energy Build. 2011, 43, 3680–3684. 10. Marco, S.; Marta, G.; Massimo, S. Modeling and techno-economic analysis of the integration of a FC based micro-CHP system for residential application with a heat pump. Energy 2017, 120, 262–275. 11. Ramadhani, F.; Hussain, M.A.; Mokhlis, H.; Fazly, M.; Ali, J.M. Evaluation of solid oxide fuel cell based polygeneration system in residential areas integrating with electric charging and hydrogen fueling stations for vehicles. Appl. Energy 2019, 238, 1373–1388. [CrossRef] 12. Takahide, H.; Yusuke, O.; Takashi, I.; Atsushi, A. Technological assessment of residential fuel cells using hydrogen supply systems for fuel cell vehicles. Int. J. Hydrogen Energy 2017, 42, 26377–26388. 13. Kazunori, N.; Joshua, D.; Michael, E.W. Assessment of primary energy consumption, carbon dioxide emissions, and peak electric load for a residential fuel cell using empirical natural gas and electricity use profiles. Energy Build. 2018, 178, 242–253. 14. Van Biert, L.; Godjevac, M.; Visser, K.; Aravind, P.V. Dynamic modelling of a direct internal reforming solid oxide fuel cell stack based on single cell experiments. Appl. Energy 2019, 250, 976–990. [CrossRef] 15. Marco, G.; Dario, M.; Marco, S.; Cesare, P.; Siu, F.A. A versatile computational tool for model-based design, control and diagnosis of a generic Solid Oxide Fuel Cell Integrated Stack Module. Energy Convers. Manag. 2018, 171, 1514–1528. 16. Weidner, E.; Ortiz, C.R.; Davies, J. Global Development of Large Capacity Stationary Fuel Cells; JCR Technical Reports; Publications Office of the European Union: Brussels, Belgium, 2019. 17. Kim, J.H.; Kim, H.J.; Yoo, S.H. Willingness to pay fuel-cell electric vehicles in South Korea. Energy 2019, 174, 497–502. [CrossRef] 18. Eunji, Y.; Myoung, K.; Han, H.S. Well-to-wheel analysis of hydrogen fuel-cell electric vehicle in Korea. Int. J. Hydrogen Energy 2018, 43, 9267–9278. 19. Lim, S.Y.; Kim, H.J.; Yoo, S.H. Household willingness to pay for expanding fuel cell power generation in

Korea: A view from CO2 emissions reduction. Renew. Sustain. Energy Rev. 2018, 81, 242–249. [CrossRef] 20. Giuseppe, D.; Piero, B.; Erasmo, M.; Francesco, P.; Fabio, M.; Dawid, P.H.; Vasilije, M. Feasibility of CaO/CuO/NiO sorption-enhanced steam methane reforming integrated with solid-oxide fuel cell for near-zero-CO2 emissions cogeneration system. Appl. Energy 2018, 230, 241–256. 21. Rao, M.; Fernandes, A.; Pronk, P.; Aravind, P.V. Design, modelling and techno-economic analysis of a solid oxide fuel cell-gas turbine system with CO2 capture fueled by gases from steel industry. Appl. Therm. Eng. 2019, 148, 1258–1270. [CrossRef] 22. Koki, Y.; Akihito, O. Study on Performance Evaluation of Residential SOFC Co-Generation Systems Part 2 Equipment characteristics and installation eﬀect. Soc. Heat. Air-Cond. Sanit. Eng. Jpn. 2013, 70, 281–284. 23. Koki, Y.; Akihito, O.; Myonghyang, L. Study on Performance Evaluation of Residential SOFC Co-Generation Systems. Soc. Heat. Air-Cond. Sanit. Eng. Jpn. 2014, 203, 25–33. 24. Akito, O.; Yuki, K. Performance of residential fuel-cell-combined heat and power systems for various household types in Japan. Int. J. Hydrogen Energy 2018, 43, 15412–15422. 25. Sumiyoshi, D. The examination for suitable spec of solid oxide fuel cell co-generation system and the proposal of tank minimization by using Bas-Tab in collective housing. J. Environ. Eng. 2015, 80, 441–450. [CrossRef] Energies 2019, 12, 3893 20 of 20

26. Takahiro, Y.; Koshiro, A.; Daisuke, S. Estimation energy saving effects of sharing fuel cell with two households by simulation program: Study on effective usage of the home fuel cell in the apartment housing. J. Environ. Eng. 2018, 83, 365–374. 27. Takahiro, Y. Analysis of Impact to Applying Residential Fuel Cell by Different Lifestyle. Ph.D. Thesis, Department of Architecture of Kyushu University, Fukuoka, Japan, 2018. 28. Taehee, L.; JinHyeok, C.; TaeSung, P. Development and Performance Test of SOFC Co-generation System for RPG. In Proceedings of the Korean Society for New and Renewable Energy, Seoul, Korea, 25 June 2009; pp. 361–364. 29. Zhou, X.; Hong, T.; Yan, D. Comparison of Building Energy Modeling Programs: HVAC Systems; Lawrence Berkeley National Lab. (LBNL): Berkeley, CA, USA, 2013. 30. Han, C.; Inhee, L. A case study of decreasing environment pollution caused by energy consumption of a dormitory building which only using electricity by efficiently simulating applying residential SOFC (solid oxide fuel cell). J. Archit. Res. Archit. Inst. Korea 2019, 21, 21–29. 31. American Society of Heating, Refrigeration, and Air-Conditioning Engineers. Energy Efficient Design for Low-Rise Residential Buildings; ASHRAE: Atlanta, GA, USA, 2010. 32. Jehean, S. The Analysis of Environmental Impact Index Using Life Cycle Assessment for Korean Traditional Building. Ph.D. Thesis, Department of Architecture of the Pusan National University, Busan, Korea, 2017. 33. Cheolyong, J.; Byunglip, A.; Chihoon, K.; Wonhwa, H. A Study on the insulation performance of the super window considering the evaluation of building energy rating. J. Korean Sol. Energy Soc. 2009, 29, 1598–6411. 34. Kim, C.; Ahn, B.; Kim, J.; Jang, C. A Study on the Evaluation of Building Energy Rating Considering the Insulation Performance of the Window and Wall in Apartment Houses. In Proceedings of the Conference of The Society of Air-conditioning and Refrigerating Engineers of Korea, Seoul, Korea, 25 June 2009. 35. Korea Environmental Industry & Technology Institute. Available online: www.epd.or.kr (accessed on 6 August 2018). 36. Arata, N.; Zacharie, W.; Patrick, M.; Stefan, D.; Gunter, S.; Jan, V.; Daniel, F. Electrochemical model of solid oxide fuel cell for simulation at the stack scale. Part I: Calibration procedure on experimental data. J. Electrochem. Soc. 2011, 158, B1102–B1118. 37. Marcin, M.; Katarzyna, B.; Shinji, K.; Janusz, S.S.; Grzegorz, B. A Multiscale Approach to the Numerical Simulation of the Solid Oxide Fuel Cell. Catalysts 2019, 253, 1–28. 38. Kwang, H.L.; Richard, K.S.; Shinji, K. SOFC cogeneration system for building applications, part 1: Development of SOFC system-level model and the parametric study. Renew. Energy 2009, 34, 2831–2838.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).